YJYujae Jeon

Yujae Jeon

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yujae Jeon has a total of 9 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Constrained Bayesian Updating Based on Posterior Predictive Checks for Reverse Osmosis Membrane Maintenance in Wastewater Treatment

    by Duc Anh Nguyen, Yujae Jeon, Viet Bac Nguyen, Hongrae Im, Taehun Kim, Am Jang

    Reliable prediction of remaining useful life (RUL) is crucial for the predictive maintenance of RO membranes in semiconductor wastewater treatment. However, existing prognostic methods often struggle with the complex nonlinear fouling dynamics, periodic cleaning resets, and noisy operating conditions typical of industrial membrane systems. This study provides a predictive foundation for a real-time early warning system for RO maintenance, utilizing a framework combining Sparse Polynomial Identification (SPI) with a Constrained Bayesian Update (CBU) method. The SPI model captures nonlinear membrane degradation through a limited set of power basis functions, and Sparse Bayesian Learning automatically selects the most relevant functions. For real-time predictions, the CBU method employs Posterior Predictive Check (PPC) to assess the reliability of predictions, while Gaussian Mutation enhances particle diversity when reliability decreases. This approach prevents the posterior drift commonly seen in standard Bayesian updating (TBU). The system was developed and validated using 17,880 hourly data points from a full-scale semiconductor wastewater treatment plant, covering two independent RO trains. A multitask architecture addresses two complementary temporal scales: RUL prediction for cleaning cycles (RMSE=81.4, a 43% to 1921% accuracy improvement over traditional methods) and membrane lifetime (RMSE=238.0, 110% to 2421% improvement). When benchmarked against TBU, linear regression, Long-Short-Term-Memory networks, and exponential degradation models, the CBU method consistently demonstrated superior accuracy, better uncertainty calibration, and enhanced computational stability. The GM mechanism keeps PPC acceptance rates close to 1.0, effectively preventing catastrophic posterior collapse amid high inter-cycle variability. This framework relies solely on routine sensor data and is easily transferable to other periodically membrane systems.

  • Open Access

    Interpretable artificial intelligence models for reverse osmosis optimization in semiconductor wastewater treatment: From reactive to proactive operation

    by Yujae Jeon, Duc Anh Nguyen, Hongrae Im, Hyewon Jeon, Am Jang

    Reverse Osmosis (RO) membranes are essential for wastewater polishing in semiconductor manufacturing; however, their operational optimization is hindered by reactive maintenance strategies, limited process transparency, and the lack of data-driven operating guidelines tailored to the specific fouling dynamics of semiconductor wastewater. This study introduces an interpretable artificial intelligence framework for RO membrane optimization, developed using 17,880 hours of operational data from a full-scale semiconductor wastewater treatment plant. Fourteen machine learning models across five algorithmic families (linear, support vector machine, instance-based, ensemble tree, and neural network) were benchmarked following an automated six-method progressive data cleaning (80.3% retention). Extra Trees achieved the best predictive performance (R2 = 0.973, RMSE = 0.234 bar for TMP; R2 = 0.999, RMSE = 0.257 m3/h for permeate flow), validated on an independent RO stream. Multi-model SHAP analysis indicated that membrane age and feed flow rate are the primary predictors of TMP and permeate flow, respectively, while revealing model-dependent biases: linear models misranked feed flow above membrane age for TMP because they fail to capture non-linear fouling accumulation. A Monte Carlo feasibility sampling framework (500,000 evaluations) identified 0.98% of operating conditions that meet simultaneous targets for TMP (≤4.38 bar) and permeate flow (≥207 m3/h), with density-based preferred zones defining robust operating windows: comprehensive chemical cleaning before day 250, a feed flow at 258–259 m3/h, and TOC levels below 4,600 ppb. This integrated framework shifts RO operation from reactive to proactive management, making it directly applicable to other membrane applications with sufficient operational data.

  • Open Access

    An interpretable artificial intelligence framework for defining the operational basin of full-scale membrane bioreactors in semiconductor wastewater treatment

    by Yujae Jeon, Duc Anh Nguyen, Kim Anh Thi Nguyen, Quoc Thai Nong Quoc Thai Nong, Am Jang

    This study presents an interpretable machine learning framework for optimizing the operation of a full-scale membrane bioreactor that treats semiconductor fabrication wastewater. We systematically evaluated 16 models using six major algorithms on 4,593 hourly observations, which included seven process inputs and three simultaneous targets: transmembrane pressure (TMP), permeate flow, and membrane tank water level. The Extra Trees model achieved the highest accuracy across all outputs, with R2 values of 0.988, 0.933, and 0.908, and corresponding RMSE values of 0.010, 0.066, and 0.341, as well as MAE values of 0.005, 0.052, and 0.227, respectively. A multi-method interpretability analysis, combining SHAP and Pearson correlation, revealed that SRT is the dominant driver of TMP, with increasing SRT within this facility's extended aeration range (43–108 days) progressively worsening fouling through biomass accumulation and cake compaction; the aeration rate exhibits a non-monotonic optimum for TMP near 5,500–6,000 m3/h with diminishing returns at higher rates, while inversely affecting permeate flow through the plant's control-coupled hydraulic response; and the F/M ratio serves as the critical lever for near-term level control. A Monte Carlo feasibility analysis applied simultaneous performance constraints to 107,848 operating points (out of 500,000 candidates). This revealed that the optimal zone is a density-concentrated core that is substantially narrower than the technically feasible envelope: SRT 50–70 days, HRT 6.0–7.0 h, and F/M <0.040 day⁻1. Reapplying SHAP within the feasible region revealed that the C/N ratio emerges as the second-ranked TMP driver, a semiconductor-specific finding reflecting denitrification-coupled organic loading from fabrication discharges. This framework offers the first evidence-based, multi-objective operating protocol for industrial semiconductor MBR systems, derived entirely from historical operational data, and is directly transferable to other membrane applications with sufficient operational data.

  • Open Access

    Super-sorption capacity and multi-cycle recovery of urea from aqueous solutions by vesicle-like iron phosphate-based carbon nano-adsorbent

    by Am Jang, Anh Nguyen, Yujae Jeon, Kim-Anh Thi Nguyen

    Despite the versatile potential applications of urea, its utilization is hindered by its unfavorable characteristics for conventional treatment methods.This necessitates the development of urea removal and recovery methods. Therefore, this study developed vesicle-like iron phosphate-based carbon (IP@C400) as a breakthrough urea removal and recovery adsorbent. IP@C400 exhibited an exceptional adsorption capacity (2242 mg/g) at a fast rate (within 2 h) across a wide range of pH environments, even in the presence of high concentrations of co-existing components, compared with the top 60 recent urea adsorbents. The adsorption process followed dual pseudo-kinetic and Langmuir isotherm models, indicating the involvement of both chemical and physical interactions between urea and the homogeneous monolayered adsorbent surface. The mechanisms included H-bonding and electrostatic attraction, as demonstrated by FT-IR and pH investigations. Remarkably, IP@C400 can be reused 10 times without significant loss of efficiency (90%) or recovery (95%)–a significant improvement over disposable options. The tests also showed safe leaching of iron and phosphorus, with levels below drinking water standards. IP@C400 worked effectively on real-world water samples, resulting in the complete removal and excellent recovery of urea. Thus, IP@C400 affords a sustainable, high-performance solution to the critical ultrapure water issue that industries must urgently address.